Applied Psychology - 2002 - Bar Eli - The Effect of Mental Training With Biofeedback On The Performance of Young Swimmers

14640597, 2002, 4, Downloaded from https://iaap-journals.onlinelibrary.wiley.com/doi/10.1111/1464-0597.00108 by INSTITUTION SUPERIEUR DE L'EDUCATION PHYSIQUE DU KEF, Wiley Online Library on [02/02/2023].
See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
APPLIED PSYCHOLOGY: AN INTERNATIONAL REVIEW, 2002, 51 (4), 567–581
The Effect of Mental Training with Biofeedback

Bar-eli etUK
Blackwell
Oxford,
Applied
APPS
©
0269-994X
1
4
51
2002
Original
128
The
01 al.
International
Effect
Psychology:
Article
Science
of Mental
Association
LtdTraining
an International
foronApplied
YoungReview
Psychology,
Swimmers 2002
on the Performance of Young Swimmers

Michael Bar-Eli*
Department of Business Administration, Ben-Gurion University of the
Negev, Beer-Sheva, Israel; Ribstein Center for Sport Medicine Sciences
and Research, Wingate Institute, Israel
Raya Dreshman
Ohalo College for Education, Tourism and Sport, Katzrin, Israel
Boris Blumenstein and Yitzhak Weinstein

Ribstein Center for Sport Medicine Sciences and Research, Wingate
Institute, Israel; Ohalo College for Education, Tourism and Sport,
Katzrin, Israel
Cette recherche porte sur la relation que l’entraînement mental entretient avec
le biofeedback et la performance. C’est une version adaptée de l’approche en
cinq étapes de Wingate qui a été utilisée comme technique de préparation
mentale en vue de l’amélioration des performances en natation chez des
nageurs de 11 à 14 ans. Les participants (N = 38) ont été distribués au hasard
à l’une des deux conditions suivantes: (a) Un entraînement expérimental
régulier assorti de trois étapes du programme Wingate; (b) Un entraînement
de controle régulier avec de la relaxation. Après une mesure de départ, les
participants ont été testés sur des scores d’évaluation et sur leur performance
réelle deux fois sur une période de quatorze semaines. Le groupe expérimental
progressa davantage, mais le groupe contrôle s’améliora aussi. Ces résultats
sont examinés à partir des travaux antérieurs sur l’approche de Wingate. En
outre, on aborde certains problèmes méthodologiques et théoriques qui con-
cernent spécialement l’appel à de telles procédures dans d’autres spécialités
sportives chez les enfants, les adolescents, et les adultes.
The purpose of the present study was to investigate the relationship between
mental training with biofeedback and performance. An adapted version of the
Wingate five-step approach was used as a mental preparation technique for
enhancing the swimming performance among 11–14-year-old children swimmers.
* Address for correspondence: Professor Michael Bar-Eli, Director, Department of Behavioral

Sciences, Ribstein Center for Sport Medicine Sciences and Research, Wingate Institute,
Netanya 42902, Israel. Email: psychwin@post.tau.ac.il
© International Association for Applied Psychology, 2002. Published by Blackwell Publishers,

108 Cowley Road, Oxford OX4 1JF, UK and 350 Main Street, Malden, MA 02148, USA.
14640597, 2002, 4, Downloaded from https://iaap-journals.onlinelibrary.wiley.com/doi/10.1111/1464-0597.00108 by INSTITUTION SUPERIEUR DE L'EDUCATION PHYSIQUE DU KEF, Wiley Online Library on [02/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
568 BAR-ELI ET AL.
Participants (n = 38) were randomly assigned to one of two conditions: (a)

experimental—regular training plus three stages of the Wingate mental train-
ing program, and (b) control—regular training and relaxing activities. After a
baseline measurement, participants were tested on evaluation scores and
actual performance twice during a 14-week period. Results indicated that the
experimental group exhibited a greater increase in performance, although the
control group also displayed some improvements. Results are discussed in
reference to previous work on the Wingate approach to mental training. In
addition, several methodological and theoretical aspects are discussed that are
particularly relevant to the use of such interventions with other athletic tasks
among children, adolescents, and adults.
INTRODUCTION
A variety of approaches to mental training and strategies for regulating
arousal have been suggested in the field of applied sport psychology (Hardy,
Jones, & Gould, 1996; Sachs, 1991; Williams, 1993). Many of these tech-
niques can be used to achieve an appropriate level of arousal by a compet-
ing athlete, thereby enhancing his or her performance. However, most
psychological skills training methods have been targeted towards adult elite
athletes (Weinberg & Williams, 1993). It has been proposed recently (e.g.
by Tremayne, 1995) that such methods should be introduced for children
and/or young athletes. For example, the performance and personal develop-
ment of young athletes who are taught psychological skills training tech-
niques may progress faster than the performance and personal development
of youngsters who do not exercise such training (Vealey, 1988). Moreover,
children given the opportunity to participate in such training programs
adjusted to their developmental level, are likely to have more effective,
enjoyable, and successful experiences in sport (Tremayne, 1995).
One of the most powerful techniques for facilitating learning of arousal
self-regulation is biofeedback (BFB). According to Zaichkowsky and
Takenaka (1993), BFB is a technique that uses instruments, usually with
sensors and transducers, which provide information regarding the state of
selected biological functions that typically are not under voluntary control.
This information is displayed to the individual. However, discussion about
biofeedback techniques in regulating arousal has been underrepresented in
the professional practice literature (Tenenbaum & Bar-Eli, 1995). In particu-
lar, an area of considerable relevance to professional practice is the use of
biofeedback as part of a larger multifaceted treatment program (Collins,
1995; Petruzzello, Landers, & Salazar, 1991). However, in general, young
athletes were neglected within this framework.
The application of BFB to sport has been widely discussed by researchers,
recognising the great potential offered by psychophysiology to the under-
standing and enhancement of athletic performance (see Sandweiss & Wolf,
© International Association for Applied Psychology, 2002.

THE EFFECT OF MENTAL TRAINING ON YOUNG SWIMMERS 569
1985; Zaichkowsky & Fuchs, 1988, 1989). The majority of these researchers
have investigated BFB effects in reducing performance anxiety, although
some have also examined the use of BFB in increasing muscle strength,
reducing pain and fatigue, increasing flexibility, and regulating heart rate.
However, there is no question that there is a need to conduct more studies
that look into the efficacy of BFB-assisted mental training interventions in
enhancing athletic performance.
Within this framework, a five-step approach to mental training incorpor-
ating BFB with videocassette recorder (VCR) has recently been presented
by Blumenstein, Bar-Eli, and Tenenbaum (1997a) for adult elite athletes.
The technique consists of five stages, with flexible time-session limits that
can be individualised. These are (a) introducing self-regulation techniques,
(b) determining and strengthening the appropriate BFB modality, (c) BFB
training with simulated competitive stress, (d) transformation of the mental
training to practice, and (e) realisation of the technique in competitive situations.
The fundamentals of this approach were presented earlier by Blumenstein,
Tenenbaum, Bar-Eli, and Pie (1995), whose two-stage mental preparation pro-
cedure consisted of using BFB apparatus and VCR equipment, combined with
relaxation and/or excitation techniques, to simulate the sensations of compet-
itive situations. Moreover, to gain scientific credibility for this approach,
Blumenstein, Bar-Eli, and Tenenbaum (1995) investigated the effects of auto-
genic training, imagery and music training on physiological indices and athletic
performance, using the BFB device. The results of this study indicated that BFB
had a substantial augmenting effect on physiological components and athletic
performance when accompanied mainly by autogenic training, and imagery.
Similar augmentation effects were revealed by Blumenstein, Breslav, Bar-Eli,
Tenenbaum, and Weinstein (1995). However, in order to enable practitioners
to successfully use the proposed method, a more detailed description of its
various stages was required. To meet this requirement, the (Wingate) five-step
approach was later developed by Blumenstein et al. (1997a).
Essentially, the five-step approach requires specific adaptations, in order
to be effectively applied in various sport disciplines and/or with different
populations (Blumenstein et al., 1997a). For example, adapted versions of
the five-step approach were specifically developed for sport disciplines such
as judo and wrestling (Blumenstein, Bar-Eli, & Tenenbaum, 1997b) and
canoe/kayak (Blumenstein & Bar-Eli, 1998). However, no investigation has
been conducted thus far in swimming. In addition, this approach was not
tested with children, who make up quite a high proportion of those particip-
ating in sport in general and swimming in particular. Therefore, this study
made an initial attempt to examine the effectiveness of this approach with
child swimmers, using an adapted version which includes only three steps.
Swimming is an individual sport, which requires that athletes compete
mainly “against themselves”, as opposed to combat sport such as judo or

570 BAR-ELI ET AL.
wrestling, in which the five-step approach was studied previously. In addi-

tion, swimmers usually begin to specialise in their sport at a much younger
age in comparison to many other sports (including canoe/kayak). Therefore,
in this study it was decided to focus on young swimmers as participants in
a program which was based on the Wingate five-step approach. In addition,
it should be noted that at face value, this approach seems to have a special
appeal for adolescents, because it is sensitive to the relatively large interindi-
vidual variance in development which characterises puberty.
One of the primary purposes of the training process with child swimmers
is the acquisition of the technical skills required for adequate performance.
In order to apply the present approach for this purpose, we hypothesised
that it would be sufficient to operate only the first three steps, namely,
introduction, identification, and simulation. These steps are connected more
to the instructional, coach–child athlete process, whereas transformation
and realisation (i.e. the last two steps) are more appropriate for elite athletes
preparing for competition.
METHOD
Participants and Design

Participants were 38 male (n = 31) and female (n = 7) Israeli swimmers, aged
11 to 14 years (M = 12.84, SD = 0.88), residing in Haifa (the third largest
city in Israel). The children trained regularly for at least two years, in one
of the two branches of the same local club, and participated in 3–4 competi-
tions a year (on a local, regional, or national basis).
The club and children voluntarily participated in the study. Conditions
(i.e. experimental and control) were randomised between the two branches,
with each branch corresponding to only one condition in order to reduce the
possibility of exchange of communication among participants from different
groups and to reduce social comparison. In other words, branches, rather
than participants, were randomly assigned to either the experimental or the
control conditions. It should be noted that the branches were located in two
opposite parts of the city to further minimise the possibility of social com-
parison. Moreover, the coaches and training programs for the two branches
were identical, because the club insisted on the same training environment
in all its branches. Thus, any potential differences between experimental and
control groups could not be attributed to different coaching procedures.
The male-female distribution was similar in each of the groups (17:3 in
the experimental group and 14:4 in the control group). This distribution
roughly reflects the situation of competitive swimming all over Israel—that
is, a vast majority in favor of males. Since participants were measured three
times (baseline plus two additional measurements) over the course of a

14-week training period (see below), a 2 × 3 (Condition × Measurement)
design was used, with repeated measures in the Measurement factor.
Treatment Conditions, Dependent Variables,

Procedure, and Instrumentation
The experimental and control groups met 38 times each over the course of
14 weeks. In the first, baseline session (first week), each of the participants was
requested to swim 50 m freestyle as fast as he or she could. The participants
were swimming individually, to eliminate the element of direct competition.
In this session, the participants were evaluated concerning their swimming
technique. The evaluations were conducted for each participant by three of
Israel’s top swimming coaches independently, using the “classification of
typical mistakes in freestyle” sheet (see below), especially developed for this
purpose. In addition, the participants’ swimming time was measured.
In sessions 2–17, which took place during weeks 2–7 of the study (basically,
three sessions per week, approximately every second day), the experimental
group underwent step 1 of the treatment. In this step (labeled “Introduc-
tion”, see Blumenstein et al., 1997a), the purpose was to teach the athletes
various self-regulation techniques, such as autogenic training, imagery, and
BFB in a “laboratory” setting (in our case, in the locker room of the club).
Each group session lasted for about 35 minutes. This treatment was given
to the experimental group in addition to their regular training sessions.
The self-regulation techniques taught during this period (i.e. sessions 2–17)
are described in more detail elsewhere (Blumenstein et al., 1997a, pp. 443–
444). However, to suit these techniques to the specific needs of our par-
ticipants (i.e. child swimmers), several minor adaptations were made. For
example, the participants were asked to imagine themselves swimming 50 m
freestyle, performing the “ideal” technique required by the coach; each
swimmer was instructed to emphasise the correction of his or her typical
own (previous) technical mistakes. During this period, imagined swimming
took place under tension-free training conditions, whereas during the later
steps (“Identification” and “Simulation”, see below) the imagined swimming
was performed in tense competitive situations. BFB was used strictly in line
with the instructions of the “Five-Step Approach” (Blumenstein et al.,
1997a), except that the children in our study were introduced to the BFB
channels in the following order: HR, EMG, GSR. The reason for this par-
ticular adaptation was that swimmers regularly use HR information to con-
trol their training load and rest between training sessions. Therefore, it was
decided to start with HR in our particular case, according to an important
principle used by the “Five-Step Approach”, namely that of “building on
what athletes already do” when systematically teaching them psychoregulat-
ive techniques (Allmer, 1981).

572 BAR-ELI ET AL.
During this period (sessions 2–17), the control group underwent ident-
ical regular training sessions to prevent any potential influence of different
training environments. However, to exclude the “Hawthorne Effect” as a
potential explanation for any possible performance improvements in the
experimental group, the control group underwent—also for about 35 min-
utes each session—various relaxing activities such as listening to quiet
music, observing nature movies, and playing table games. This procedure
would enable any potential performance improvements in the experimental
group—in comparison to the control group—to be attributed to the effect
of the treatment (i.e. mental training incorporating BFB).
In session 18, both groups were measured in a procedure which was
identical to the one they underwent during the first, baseline session. Thus,
they were requested to individually swim 50 m freestyle, and were again
rated on the “classification of typical mistakes in freestyle” sheet by the
same coaches; in addition, their swimming times were again measured.
In sessions 19–37, which took place during weeks 7–13 of the study
(again, three sessions per week, every second day), the experimental group
underwent steps 2 and 3 of the treatment (see Blumenstein et al., 1997a). In
step 2 (labeled “Identification”), the athletes’ purpose was to identify and
strengthen the most efficient BFB response modality, which could be HR,
EMG, or GSR. This process was conducted using the test of self-regulation
(SRT) of arousal, which is used to examine athletes’ baseline self-regulation
levels before applying the entire program and later, between the various
steps of the program (for further details concerning the SRT, see Blumen-
stein et al., 1997a, pp. 444–445). The identification process was aided by the
sport psychologist and was based on the athlete’s responses (HR, EMG, or
GSR) reflected on the (portable) computer’s screen. In step 3 (labeled “Sim-
ulation”), the athletes’ purpose was to conduct BFB training with simulated
competitive stress, using auditory, but not visual VCR, simulations (see
below). Both steps were conducted in the “laboratory” (i.e. locker room).
Each group session lasted about 35 minutes. Again, this treatment was
given to the experimental group in addition to their regular training.
It should be noted that whereas the Identification step was completed
(that is, lasted 15 sessions), only 4 sessions (and not 15) took place in the
Simulation step applied in this study. The reason for this adaptation was
the quite limited competitive experience of the children participating in the
study, which could have rendered the use of self-presentational visual VCR
less effective. Therefore, only auditory simulation was used for a shorter
period (i.e. 4 sessions) in this step. The auditory simulation consisted of
original (competitive) noise, recorded during a high-level international
swimming contest that took place earlier in Israel.
The Simulation step might have provided basic learning about what to ex-
pect, since the swimmers in our study were quite inexperienced in competition.

In other words, participants in the experimental group not only got a
chance to generalise any self-regulatory learning that they had, but they also
had a chance to learn about competition—an opportunity that the other
group did not have, which may prove to be a particularly efficient tool with
children. During this period (that is, sessions 19–37), the control group con-
tinued to be treated as mentioned above. In session 38, both groups were
measured again in a procedure which was identical to the one they under-
went during the two previous measurements.
In order to evaluate the swimmers’ technique, the three experts developed
a “classification of typical mistakes in freestyle” sheet, which consisted of
various technical elements belonging to one of the following categories: (a)
Body position in the water (not horizontal, too tensed, excessive trunk’s
angle); (b) Head position (turned sideward too much, excessively high
head position, excessively low head position in line with spinal cord); (c)
Armstroke performance (hand too far from body, elbow too low, elbow is
unlocked at the end of stroke); (d) Legstroke performance (toes are relaxed
and point outward, knees are too straight or overbent, kick starts at knee
instead of pelvis); (e) Breathing (inhaling with tilt angle too high, opposite
hand with head turn unsynchronised, inhaling at one side only). A full
agreement between the expert coaches existed as to the relevance of these
categories and elements for the evaluation of swimmers’ freestyle technique.
Each of the 15 elements included in this sheet was evaluated on a Likert
scale ranging from extremely incorrect (1) to extremely correct (5). A swim-
mer’s final evaluation in a particular measurement session consisted of the
mean grade calculated across all 15 elements. It should be noted that in the
judgment and decision making literature, holistic ratings are usually con-
sidered less reliable and valid in comparison to judgments which are decom-
posed into their specific elements and later recomposed, as was done here
(see Tenenbaum & Bar-Eli, 1993). Accordingly, the internal consistency (0.92)
and the mean interrater agreement rate (91.9%) were satisfactory.
It was important to measure the swimmers’ performance for several
reasons. First, a reduction in a swimmer’s mistakes should obviously be
reflected in his or her performance. Thus, performance (i.e. swimming time
in seconds) was measured to validate experts’ evaluations. Moreover, it
could be argued that since the three expert coaches were aware of the
study’s hypothesis (that is, they knew that the experimental group was
expected to demonstrate a greater improvement rate in comparison to the
control group), their evaluations could have been biased accordingly. To
control for this possibility, we made sure that the coaches who conducted
the experimental and control conditions were unaware (blind) as to the real
purpose and hypothesis of the study. Hence, performance provided an
objective, unbiased measure in this respect as well.
A schematic description of the entire research design is presented in Table 1.

574
TABLE 1
BAR-ELI ET AL.
Schematic Description of the Research Design
Session (35 min each)
1 (Baseline) 2–17 18 19–37 38

Group (week 1) (weeks 2–7) (week 7) (weeks 7–13) (week 14)
Experimental • 50 m freestyle • step 1: • 50 m freestyle • step 2: Identification • 50 m freestyle

(n = 20) • technique Introduction • technique evaluation (complete); step 3: • technique evaluation
evaluation (complete) Simulation (partial)
• regular training • regular training

Control • 50 m freestyle • relaxing activities • 50 m freestyle • relaxing activities • 50 m freestyle
(n = 18) • technique • regular training • technique evaluation • regular training • technique evaluation
evaluation
RESULTS
The Bartlett probability values (chi-square tests), sensitive to non-normality
(Bartlett, 1937; Kirk, 1968), indicated no violation of the homogeneity of
variance assumption in any of the conditions, for both evaluation scores
and performance (swimming times). Furthermore, the interaction between
the baseline measures (covariates) and the experimental/control conditions was
not significant. Thus, the assumptions underlying the subsequent repeated-
measures analyses were satisfactorily met.
A one-way analysis of variance was conducted on baseline (measurement
1) evaluation scores to determine if an initial difference existed between the
experimental/control conditions. Results indicated no significant difference
among the groups in the outset of the study. Thus, a 2 × 3 (Condition ×
Measurement) mixed-factorial ANOVA was performed, with Condition
representing a between-subjects factor and Measurement representing a
within-subjects factor.
The analysis revealed a significant interaction, Condition × Measurement,
F(1,35) = 9.27, P < 0.004. The effect size (ω2) was 0.28. Means and standard
deviations are presented in Table 2.
The Condition × Measurement effect is presented in Fig. 1. The sig-
nificant Condition × Measurement interaction was attributed mainly to the
greater rate of progress obtained by the experimental group throughout the
study, in comparison to the control group, as indicated by the slopes in
Fig. 1. This is evident also from the increasing differences between the two
groups at each measurement, from the nonsignificant difference at baseline,
through the 2nd measurement, F (1,36) = 6.89, P < 0.013, to the largest
difference obtained at the 3rd measurement, F(1,36) = 41.50, P < 0.000. The
effect sizes (strength of association calculated employing ω2 estimates) were
0.02, 0.13, and 0.52, respectively.
An analysis of the simple main effects of time for each group was con-
ducted to indicate whether the groups developed differently over time.
TABLE 2
Means and Standard Deviations for Evaluation Scores by Measurements
Measurement
Group 1 (Baseline) 2 3
Experimental (n = 20)
M 2.97 3.55 4.42
SD 0.32 0.27 0.23
Control (n = 18)
M 2.90 3.31 3.75
SD 0.38 0.28 0.40

576 BAR-ELI ET AL.
FIGURE 1. Evaluation scores for experimental and control groups across

measurements.
Within the experimental group, the difference between time 1 and time 2 was
significant, F(2,36) = 67.98, P < 0.000, and so was the difference between
time 2 and time 3, F(2,36) = 153.33, P < 0.000. The effect sizes (ω2) were
0.61 and 0.85, respectively. Within the control group, these differences were
also significant, F(2,36) = 21.93, P < 0.000 and F(2,36) = 30.32, P < 0.000,
with effect sizes (ω2 estimates) of 0.35 and 0.42, respectively.
A one-way analysis of variance was conducted on baseline (measurement
1) performances (swimming times in seconds), to determine if an initial dif-
ference existed between the experimental and control conditions. Results
indicated a significant, F(1,36) = 19.09, P < 0.000, difference between the
groups in the outset of the study (ω2 = 0.32). Thus, a 2 × 3 (Condition ×
Measurement) mixed-factorial ANCOVA was performed, with Condition
representing a between-subjects factor, Measurement representing a within-
subjects factor, and performances (swimming times) in baseline as a covariate.
The analysis revealed a significant interaction, Condition × Measurement,
F(2,70) = 12.07, P < 0.000 (ω2 = 0.32). Means and standard deviations are
presented in Table 3.
The Condition × Measurement effect is presented in Fig. 2. The signific-
ant Condition × Measurement interaction was attributed mainly to the
improvement in performance (i.e. consistent decrease in swimming times)
obtained by the experimental group throughout the study, in comparison
to the relatively stable performances (swimming times) of the control group,
as indicated by the slopes in Fig. 2. This is evident also from the increasing

TABLE 3
Means and Standard Deviations for Performances
(Swimming Times in Seconds) by Measurements
Measurement
Group 1 (Baseline) 2 3
Experimental (n = 20)
M 33.36 32.10 30.99
SD 2.87 2.81 2.48
Control (n = 18)
M 38.29 38.37 37.61
SD 4.04 4.07 4.26
FIGURE 2. Performances (swimming times in seconds) for experimental and

control groups across measurements.
differences between the two groups at each measurement, from baseline (see
above) through the 2nd Measurement, F(1,36) = 29.90, P < 0.000 (ω2 = 0.44),
to the largest difference obtained at the 3rd measurement, F(1,36) = 37.98,
P < 0.000 (ω2 = 0.47). Evidently, as a whole, the performance of the experi-
mental group was significantly better, F(1,35) = 95.39, P < 0.000 (ω2 = 0.68),
with swimming times substantially lower than those of the control group
(M = 30.99, SD = 2.48 vs. M = 37.61, SD = 4.26, respectively).
An analysis of the simple main effects of time for each group was con-
ducted to indicate whether the groups developed differently over time.
Within the experimental group, the difference between time 1 and time 2

578 BAR-ELI ET AL.
was significant, F(2,36) = 38.46, P < 0.000, and so was the difference
between time 2 and 3, F(2,36) = 23.66, P < 0.000. The effect sizes (ω2) were
0.48 and 0.38, respectively. Within the control group, only the difference
between time 2 and time 3 was significant, F(2,36) = 6.58, P < 0.004,
whereas the difference between time 1 and time 2 was nonsignificant (ω2
estimates of 0.18 and 0.02, respectively). Finally, it should be noted that the
correlation between speed and rating (0.52) was significant (P < 0.000).
DISCUSSION
The purpose of the present investigation was to further test the relationship
between mental training with BFB and the performance of child swimmers,
while trying to control for social comparison. Accordingly, the present
investigation was designed to determine whether using three of the five steps
proposed by Blumenstein et al. (1997a) would lead to better results—in
terms of expert coaches’ evaluations and actual performance—than using
regular training alone, with child swimmers.
Results from both measures (i.e. dependent variables) indicated that
participants in the experimental group consistently exhibited the greatest
improvements. In other words, it can be concluded that using regular train-
ing plus mental training with BFB (in line with Blumenstein et al.’s 1997a
approach) will lead to better results than using regular training only (while
controlling for the “Hawthorne effect”). This conclusion is justified by the
significant interactions obtained for both dependent variables in this invest-
igation, which indicate not only that using our mental training approach
leads to better results than using regular training alone, but also that the
superiority of this approach increases over time (measurements). These
results validate our basic hypothesis, derived from both previous research
(Blumenstein, Bar-Eli et al., 1995; Blumenstein, Breslav et al., 1995) and
application (Blumenstein & Bar-Eli, 1998; Blumenstein et al., 1997a, 1997b;
Blumenstein, Tenenbaum et al., 1995) related to the Wingate five-step
approach.
It could be argued that the performance results (swimming time in sec-
onds) are contaminated by a ceiling effect. In other words, it is possible that
the experimental group had less room for improvement, due to its signific-
antly better performance (faster swimming times) at baseline. However,
such a possible ceiling effect would act against our hypothesis, that is, would
make the exploration of any treatment effects more difficult. Therefore, the
interactional effect revealed for this dependent measure, which was attrib-
uted mainly to a consistent, substantial decrease in swimming times in the
experimental group, in comparison with the relatively stable performances
of the control group, makes the effect of the present mental training
approach even more impressive.

In the present investigation each group was randomly assigned to the
experimental or control condition, which effectively prevented any compar-
ison of performance or competition by the control group with participants
in the experimental condition. For example, this eliminated the possibility
of participants in the experimental treatment group encouraging particip-
ants in the control group. It is reasonable to assume that this precaution
was important in controlling competition and social comparison effects, and
thus, results concurred with our previous research, demonstrating the effec-
tiveness of this mental training approach in enhancing swimming perform-
ance. It appears that when social comparison processes and competition
are controlled for, this approach produces the best performance, although
regular training alone is naturally also effective in enhancing performance.
However, such an approach is effective only when one makes sure that there
is an identical training/coaching environment in both groups (as was done
in this study), in order to control for any contaminating effects.
For future research, it is recommended that the effectiveness of the Win-
gate five-step approach be investigated in other specific populations (e.g.
adolescents) and sport disciplines (e.g. judo, wrestling, fencing), including
team sports. This recommendation is based on the fact that the present
program must be specifically tailored to the major characteristics of the
athlete under consideration and his or her sport discipline (Blumenstein
et al., 1997a). In such cases, it may be that all participants should receive
treatment believed to be effective; then, the design and conduct of “quasi-
experiments” is recommended, particularly in field settings (Cook &
Campbell, 1976, 1979).
It should be noted that in this study, participants were assigned to condi-
tions as a whole group rather than as individuals. Accordingly, the present
investigation could also have been viewed as a “quasi-experiment”. However,
similar designs are often considered in different areas of applied sport and
exercise psychology as “true” experiments; moreover, “quasi”-experimental
designs often permit reasonable causal inferences even when an investiga-
tion is not “truly” experimental in nature. In addition to the nomothetic-
quantitative methods employed in this study, ideographic (Bryan, 1987;
Smith, 1988) and qualitative (Martens, 1987; Patton, 1980) methods should
also be used, such as case studies and applied behavior analyses. These
research methods seem to be promising, because future research needs to
examine unique personality and situational variables that help determine an
individual’s response to being assigned to a specific treatment condition.
This is true in particular for children, due to possible large developmental
differences among them.
In summary, the findings of the present investigation provide further
support that the present approach can produce significant improvements
in athletic performance in young swimmers. As such, these findings are

580 BAR-ELI ET AL.
important because they empirically demonstrate the possibility of success-

fully and effectively applying mental training incorporating BFB. It remains
to be seen whether similar performance enhancements could be achieved
through applying similar intervention programs in other athletic settings as
well. As knowledge of the effects and limitations of such intervention pro-
grams is accrued, the credibility of these programs will be enhanced.
REFERENCES
Allmer, H. (1981). Psychologische Aspekte sportlicher Beanspruchung [Psycholo-
gical aspects of athletic strain]. In J.R. Nitsch (Ed.), Stress (pp. 503–545). Bern,
Switzerland: Huber.
Bartlett, M.S. (1937). Properties of sufficiency and statistical tests. Proceedings of the
Royal Society, 160, 268–282.
Blumenstein, B., & Bar-Eli, M. (1998). Self-regulation training with biofeedback in
elite canoeists and kayakers. In V. Issurin (Ed.), Science and practice of canoe/
kayak high performance training ( pp. 124 –132). Elite Sport Department of Israel,
Wingate Institute, Tel Aviv, Israel.
Blumenstein, B., Bar-Eli, M., & Tenenbaum, G. (1995). The augmenting role of
biofeedback: Effects of autogenic, imagery, and music training on physiological
indices and athletic performance. Journal of Sports Sciences, 13, 343–354.
Blumenstein, B., Bar-Eli, M., & Tenenbaum, G. (1997a). A five-step approach to
mental training incorporating biofeedback. The Sport Psychologist, 11, 440–453.
Blumenstein, B., Bar-Eli, M., & Tenenbaum, G. (1997b). Mental training in elite
sport incorporating biofeedback. In R. Lidor & M. Bar-Eli (Eds.), Proceedings of
the IX World Congress of Sport Psychology, Innovations in sport psychology: Link-
ing theory and practice, Part I ( pp. 130 –132). Netanya, Israel: Wingate Institute.
Blumenstein, B., Breslav, I., Bar-Eli, M., Tenenbaum, G., & Weinstein, Y. (1995).
Regulation of mental states and biofeedback techniques: Effects of breathing
pattern. Biofeedback and Self-Regulation, 20, 169–183.
Blumenstein, B., Tenenbaum, G., Bar-Eli, M., & Pie, J. (1995). Mental preparation
techniques with elite athletes using computerized biofeedback and VCR. In W.K.
Simpson, A.D. Le Unes, & J.S. Picou (Eds.), Applied research in coaching and
athletics annual ( pp. 1–15). Boston, MA: American Press.
Bryan, A.J. (1987). Single-subject designs for evaluation of sport psychology inter-
ventions. The Sport Psychologist, 1, 283–292.
Collins, D. (1995). Psychophysiology and sport performance. In S.J.H. Biddle (Ed.),
European perspectives on exercise and sport psychology (pp. 154–178). Cham-
paign, IL: Human Kinetics.
Cook, T.D., & Campbell, D.T. (1976). The design and conduct of quasi-experiments
and true experiments in field settings. In M. Dunnette (Ed.), Handbook of indus-
trial and organizational psychology ( pp. 75–176). Chicago: Rand McNally.
Cook, T.D., & Campbell, D.T. (1979). Quasi-experimentation: Design and analysis
issues for field settings. Chicago: Rand McNally.
Hardy, L., Jones, G., & Gould, D. (1996). Understanding psychological preparation
for sport: Theory and practice of elite perfrmance. New York: Wiley.

Kirk, R.E. (1968). Experimental design: Procedures for the behavioral sciences. Belmont,
CA: Brooks/Cole.
Martens, R. (1987). Science, knowledge, and sport psychology. The Sport Psycholo-
gist, 1, 29–55.
Patton, M.Q. (1980). Qualitative evaluation methods. Beverly Hills, CA: Sage.
Petruzzello, S.J., Landers, D.M., & Salazar, W. (1991). Biofeedback and sport/exer-
cise performance: Applications and limitations. Behavior Therapy, 22, 379–392.
Sachs, M.L. (1991). Reading list in applied sport psychology: Psychological skills
training. The Sport Psychologist, 5, 88–91.
Sandweiss, J., & Wolf, S. (Eds.) (1985). Biofeedback and sport science. New York:
Plenum.
Smith, R.E. (1988). The logic and design of case study research. The Sport Psycho-
logist, 2, 1–12.
Tenenbaum, G., & Bar-Eli, M. (1993). Decision-making in sport: A cognitive per-
spective. In R.N. Singer, M. Murphey, & L.K. Tennant (Eds.), Handbook of
research on sport psychology ( pp. 171–192). New York: Macmillan.
Tenenbaum, G., & Bar-Eli, M. (1995). Contemporary issues in exercise and sport
psychology research. In S.J.H. Biddle (Ed.), European perspectives on exercise and
sport psychology ( pp. 292–323). Champaign, IL: Human Kinetics.
Tremayne, P. (1995). Children and sport psychology. In T. Morris & J. Summers
(Eds.), Sport psychology, theory, applications and issues (pp. 516–537). New
York: John Wiley.
Vealey, R. (1988). Future directions in psychological skills training. The Sport Psy-
chologist, 2, 318–336.
Weinberg, R.S., & Williams, J.M. (1993). Integrating and implementing a psycho-
logical skills training program. In J.M. Williams (Ed.), Applied sport psychology
(2nd edn.) ( pp. 274 –298). Mountain View, CA: Mayfield.
Williams, J.M. (Ed.) (1993). Applied sport psychology: Personal growth to peak per-
formance (2nd edn.). Mountain View, CA: Mayfield.
Zaichkowsky, L.D., & Fuchs, C.Z. (1988). Biofeedback applications in exercise and
athletic performance. In K.B. Pandolf (Ed.), Exercise and sports sciences reviews
( pp. 381– 421). New York: Macmillan.
Zaichkowsky, L.D., & Fuchs, C.Z. (1989). Biofeedback-assisted self-regulation for
stress management in sports. In D. Hackfort & C.D. Spielberger (Eds.), Anxiety
in sports: An international perspective ( pp. 235–245). New York: Hemisphere.
Zaichkowsky, L., & Takenaka, K. (1993). Optimizing arousal level. In R.N. Singer,
M. Murphey, & L.K. Tennant (Eds.), Handbook of research on sport psychology
( pp. 511–527). New York: Macmillan.

Applied Psychology - 2002 - Bar Eli - The Effect of Mental Training With Biofeedback On The Performance of Young Swimmers

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Applied Psychology - 2002 - Bar Eli - The Effect of Mental Training With Biofeedback On The Performance of Young Swimmers

Uploaded by

Copyright:

Available Formats

14640597, 2002, 4, Downloaded from https://iaap-journals.onlinelibrary.wiley.com/doi/10.1111/1464-0597.00108 by INSTITUTION SUPERIEUR DE L'EDUCATION PHYSIQUE DU KEF, Wiley Online Library on [02/02/2023].

The Effect of Mental Training with Biofeedback

on the Performance of Young Swimmers

Boris Blumenstein and Yitzhak Weinstein

* Address for correspondence: Professor Michael Bar-Eli, Director, Department of Behavioral

© International Association for Applied Psychology, 2002. Published by Blackwell Publishers,

Participants (n = 38) were randomly assigned to one of two conditions: (a)

© International Association for Applied Psychology, 2002.

© International Association for Applied Psychology, 2002.

wrestling, in which the ﬁve-step approach was studied previously. In addi-

Participants and Design

© International Association for Applied Psychology, 2002.

Treatment Conditions, Dependent Variables,

© International Association for Applied Psychology, 2002.

© International Association for Applied Psychology, 2002.

© International Association for Applied Psychology, 2002.

Schematic Description of the Research Design

Session (35 min each)

1 (Baseline) 2–17 18 19–37 38

Experimental • 50 m freestyle • step 1: • 50 m freestyle • step 2: Identiﬁcation • 50 m freestyle

© International Association for Applied Psychology, 2002.

© International Association for Applied Psychology, 2002.

FIGURE 1. Evaluation scores for experimental and control groups across

© International Association for Applied Psychology, 2002.

FIGURE 2. Performances (swimming times in seconds) for experimental and

© International Association for Applied Psychology, 2002.

© International Association for Applied Psychology, 2002.

© International Association for Applied Psychology, 2002.

important because they empirically demonstrate the possibility of success-

© International Association for Applied Psychology, 2002.

© International Association for Applied Psychology, 2002.

You might also like